CLUSTERING WITH THE BLACKWINGED KITE ALGORITHM

Abstract

In this article, a metaheuristic optimization algorithm called the Black Kite Algorithm (BKA) is proposed, inspired by the nomadic and predatory behaviors. BKA integrates the Cauchy mutation strategy and the Leader strategy to enhance the algorithm's global search capability and convergence speed. This novel combination provides a good balance between exploring global solutions and utilizing local information.

Clustering is a widely used technique in data analysis. Its fundamental purpose is to reveal structures and relationships in a dataset by grouping data points with similar characteristics. These groups can be utilized to understand patterns in the dataset, perform data exploration, and make predictions. Clustering algorithms typically work by measuring similarities between data points using a distance metric. Determining the number of clusters is a challenging task, even when clustering is done correctly. To address these challenges, several techniques have been proposed in the literature. Most of these methods require prior knowledge of the number of clusters to be addressed, which should be provided as an algorithm parameter. Real-world clustering problems arise when the number of clusters in the data collection set is unknown in advance. The Blackwinged Kite Optimization algorithm is a powerful search algorithm that has been proposed to solve optimization problems seen as clustering problems. With this developed method, datasets are divided into clusters based on distance, and the number of clusters is also accurately determined at a satisfactory level.